The Agent Skills Directory

[SAFE]: The skill performs read-only environment checks using standard CLI tools such as git and test to orient the agent within a codebase.
[SAFE]: No network operations, sensitive file access (e.g., credentials, SSH keys), or data exfiltration patterns were identified.
[SAFE]: The instructions incorporate safety-conscious logic, explicitly requiring the agent to "orient first, then offer" and forbidding forced installation of tools or hooks without user choice.
[SAFE]: No obfuscation, prompt injection, or suspicious persistence mechanisms are present in the skill's instructions or metadata.

activation